|
Aryan Garg Simple solutions turn heads I'm a second year Computer Science PhD student at the University of Wisconsin-Madison, advised by Professor Mohit Gupta. We believe in pushing the limits of consumer grade imaging across the perception triangle: low-light, extreme motion and dynamic range. I have been fortunate to work with many brilliant professors and experts throughout my journey. Prof. Renu Rameshan ignited my interest in computer vision early during my Bachelor's at IIT Mandi which also led me to exploring generative modeling for outdoor lighting during my junior year with Prof. Jean-François Lalonde and Dr. Yannick Hold-Geoffroy. After graduating, I spent a wonderful year at IIT Madras getting closer to sensors alongside developing 3D reconstruction and generative editing methods as a Bosch AI Research Fellow at the Computational Imaging Lab directed by Prof. Kaushik Mitra. Interestingly, my vision sensors and generative interests got a human touch to them as I spent the summer of '24 at Computational Behaviour Lab, advised by Prof. Antti Oulasvirta and now Professor, Yue Jiang. Apart from research, I enjoy riding my motorcycle, baking and trekking. My previous trek (8th) was Sandakphu-Phalut in the Himalayas overlooking the majestic sleeping Buddha. I also love to travel during the holidays to meet my close friends Julian Fabinc and Noemi Ippolito. |
|
Research InterestsGenerative modeling for computer vision and imaging, with a sprinkle of human-centricity. |
|
News
|
Publications
|
Radiance Fields from Photons
Sacha Jungerman, Aryan Garg, Mohit Gupta, Transactions on Graphics, 2025 / SIGGRAPH, 2026 🇺🇸 ACM DoI / arXiv NeRF and 3D Gaussian Splatting with Photons |
|
Controllable GUI Exploration
Aryan Garg, Yue Jiang, Antti Oulasvirta Preprint, 2025 arXiv Diffusion for automatic GUI designing & ideation |
|
IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion
Tharun Anand, Aryan Garg, Kaushik Mitra WACV-W, 2025 🇺🇸 arXiv IP-FaceDiff: Identity-Preserving Facial Video Editing with Diffusion |
|
HDRSplat: Gaussian Splatting for High Dynamic Range 3D Scene Reconstruction from Raw Images
Shreyas Singh*, Aryan Garg*, Kaushik Mitra BMVC, 2024 🇬🇧 Project Page / arXiv High Dynamic Range 3D Gaussian Splatting At Night |
|
Stereo-Knowledge Distillation from dpMV to Dual Pixels for Light Field Video Reconstruction
Aryan Garg, Raghav Mallampali, Akshat Joshi, Shrisudhan Govindarajan, Kaushik Mitra ICCP, 2024 🇨🇭 GitHub / arXiv Stereo Knowledge Distillation for dp-disparity estimation, dpMV dataset, and LFVR. |
|
Parameter and Data Efficient Spectral Style DCGAN
Aryan Garg ICLR Tiny Paper,2024 Oral Presentation 🇦🇹 GitHub / arXiv / OpenReview Unconditional face generation at the speed and size of DCGAN using a relatively tiny training dataset (~4000). Convergence and higher fidelity over StyleGAN are achieved using a spectrally normalized discriminator and a tiny generator within 20 minutes on a commercial, free cloud GPU. |
|
G-PECNet: Towards a Generalizable Pedestrian Trajectory Prediction System
Aryan Garg, Renu Rameshan ICLR Tiny Paper, 2024 Oral Presentation 🇦🇹 GitHub / arXiv / OpenReview Synthetic dataset generation and sinusoidal activations lead to a winning solution over the final displacement error. |
|
Re-envisioning Sky Models
Ian J. Maquignaz, Lucas Valença, Aryan Garg, Yannick Hold Geoffroy, Julien Philip, Jean Francois Lalonde Semaine numériQC, 2023 🇨🇦 3rd Best Presentation Award SkyNet for user-controlled generation of skies incorporating the versatility of parametric models and realism from deep generative networks. |
|
DeepSky
Ian J. Maquignaz, Aryan Garg, Yannick Hold Geoffroy, Julien Philip, Jean Francois Lalonde Symposium IA Montréal, 2022 🇨🇦 Poster / Abstract DeepSky: Learning to generate photorealistic skies per user-controlled positioning of solar and atmospheric components |
Academic ServicesReviewer For:
Teaching Assistant
|
More Socials |
|
Website template was |